There is a considerable body of evidence that tumor-specific CD8+ T lymphocytes can be elicited in tumor-bearing individuals and can mediate a significant antitumor effect. In most cases, however, tumor-specific CD8 T cells do not arise naturally. We are investigating the possibility that the poor immunogenicity of some tumors is due to limitations in the abilities of cancer cells to present their tumor-specific antigens to the immune system. In the past few years, there has been tremendous progress in our understanding of how cells present antigens for recognition by CD8 T cells. It is now clear that presentation of antigen results from a chain of events, including protein breakdown into peptides of 8 to 10 amino acids, transport of peptides from the cytosol into the endoplasmic reticulum, binding of peptides to a special peptide- presenting molecule, and delivery of the peptide-complexed presenting molecule to the cell surface. To test the hypothesis that limitations in tumor antigen breakdown or transport of the antigen into the endoplasmic reticulum limit the immunogenicity of T cells, we have created novel vaccines in which antigenic peptide generation and transport occurs independently of the normal antigen processing machinery. We are currently testing whether such vaccines are able to elicit tumor-specific CD8 T cells and can protect animals against a challenge with tumor cells or cure animals with existing tumors. We have created a vaccine for possible use in patients afflicted with melanoma.